Since the outcomes of patients with candidemia is poor and Candida spp. with increased resistance to antifungal therapy may be associated with these results, the emergence of these blood infections caused by non-C. albicans Candida spp. was explored prospectively over a two-year period (2009-2010). Candidemia was defined as the recovery of Candida spp. in culture from a patient's blood sample. The in vitro susceptibility of each isolate to amphotericin B, caspofungin, fluconazole and voriconazole was determined. In addition, characteristics of patients and outcomes were investigated in real-time. The Candida distribution was compared to that observed in a similar study 10 years earlier in the same hospital. A total of 182 patients with candidemia were included in the study. While C. albicans was the most frequently isolated species (n = 102), non-C. albicans Candida spp. included; C. glabrata (n = 32), C. parapsilosis (n = 21), C. tropicalis (n = 13), C. krusei (n = 8), C. kefyr (n = 3), C. lusitaniae (n = 2), C. lipolytica (n = 2), C. famata (n = 1), C. guilliermondii (n = 1), C. inconspicua (n = 1), C. dubliniensis (n = 1), C. sake (n = 1) and C. nivariensis (n = 1). In seven patients, C. albicans was associated with another Candida spp. Surprisingly, this prospective study demonstrated that regardless of the department (intensive care unit or hematological department), Candida spp. distribution was no different from that found in the 1998-2001 survey, except for C. krusei. A reduction in the proportion of C. krusei isolates was observed from 2000-2010 (P = 0.028) as a result of its decreased recovery in the hematological department.

Background Campylobacter spp. bacteremia is a severe infection. A nationwide 5-year retrospective study was conducted to characterize its clinical features and prognostic factors. Methods Patients diagnosed with Campylobacter spp. bacteremia in 37 French hospitals participating in the surveillance network of the National Reference Center for Campylobacter and Helicobacter were included from January 1, 2015, to December 31, 2019. The goal was to analyze the effects of a delay of appropriate antibiotic therapy and other risk factors on 30-day mortality, antibiotic resistance, patient characteristics and prognosis according to the Campylobacter species. Findings Among the 592 patients, Campylobacter jejuni and Campylobacter fetus were the most commonly identified species (42.9 and 42.6%, respectively). The patients were elderly (median age 68 years), and most had underlying conditions, mainly immunodepression (43.4%), hematologic malignancies (25.9%), solid neoplasms (23%) and diabetes (22.3%). C. jejuni and Campylobacter coli were associated with gastrointestinal signs, and C. fetus was associated with secondary localizations. Among the 80 patients (13.5%) with secondary localizations, 12 had endocarditis, 38 vascular, 24 osteo-articular and 9 ascitic fluid infections. The thirty-day mortality rate was 11.7%, and an appropriate antibiotic treatment was independently associated with 30-day survival (odds ratio [OR]=0.47, 95% CI [0.24–0.93], p=0.03). The median efficient therapy initiation delay was quite short (2 days, IQR [0–4]) but it had no significant impact on 30-day mortality (p=0.78). Interpretation Campylobacter spp. bacteremia mainly occurred in elderly immunocompromised individuals with variable clinical presentations according to the species involved. Appropriate antimicrobial therapy was associated with improved 30-day survival.

Background Colistin resistance in Acinetobacter baumannii often results from mutational activation of the two-component system PmrAB and subsequent addition of phospho-ethanolamine (pEtN) to lipooligosaccharide by up-regulated pEtN transferase PmrC. Objectives To characterize mechanisms of colistin resistance independent of PmrCAB in A. baumannii. Methods Twenty-seven colistin-resistant A. baumannii were collected from 2012 to 2018. Analysis of operon pmrCAB was performed by PCR and sequencing. Seven strains were investigated further by WGS and whole-genome MLST (wgMLST). Results Seven out of the 27 selected isolates were found to overexpress eptA, a gene homologous to pmrC, likely as a consequence of upstream insertion of an ISAba1 element. Insertion sites of ISAba1 were mapped 13, 18 and 156 bp ahead of the start codon of eptA in five strains, one strain and one strain, respectively. The finding that the isolates did not cluster together when compared by wgMLST analysis supports the notion that distinct insertion events occurred in close, but different, genetic backgrounds. Conclusions Activation of eptA and subsequent addition of pEtN to the cell surface represents a novel mechanism of resistance to colistin in A. baumannii.

Objectives: The aim was to assess the incidence of sink contamination by multidrug-resistant (MDR) Pseudomonas aeruginosa and Enterobacteriaceae, risk factors for sink contamination and splashing, and their association with clinical infections in the intensive care setting. Methods: A prospective French multicentre study (1 January to 30 May 2020) including in each intensive care unit (ICU) a point-prevalence study of sink contamination, a questionnaire of risk factors for sink contamination (sink use, disinfection procedure) and splashing (visible plashes, distance and barrier between sink and bed), and a 3-month prospective infection survey. Results: Seventy-three ICUs participated in the study. In total, 50.9% (606/1191) of the sinks were contaminated by MDR bacteria: 41.0% (110/268) of the sinks used only for handwashing, 55.3% (510/923) of those used for waste disposal, 23.0% (62/269) of sinks daily bleached, 59.1% (126/213) of those daily exposed to quaternary ammonium compounds (QACs) and 62.0% (285/460) of those untreated; 459 sinks (38.5%) showed visible splashes and 30.5% (363/1191) were close to the bed (<2 m) with no barrier around the sink. MDR-associated bloodstream infection incidence rates 0.70/1000 patient days were associated with ICUs meeting three or four of these conditions, i.e. a sink contamination rate 51%, prevalence of sinks with visible splashes 14%, prevalence of sinks close to the patient's bed 21% and no daily bleach disinfection (6/30 (20.0%) of the ICUs with none, one or two factors vs. 14/28 (50.0%) of the ICUs with three or four factors; p 0.016). Discussion: Our data showed frequent and multifactorial infectious risks associated with contaminated sinks in ICUs.

No abstract